`RecordingConnection` string handling issues

[adambaratz]

I was working on test coverage for topic search (see #1252). When recording an example API response, the test runner got thrown off by a non-breaking space in the content:

Traceback (most recent call last):
  File "tests/Topic.py", line 67, in testSearch
    self.assertListKeyEqual(self.g.search_topics("python"), attrgetter("name"), expected_names)
  File "tests/Framework.py", line 312, in assertListKeyEqual
    realKeys = [key(element) for element in elements]
  File "github/PaginatedList.py", line 61, in __iter__
    newElements = self._grow()
  File "github/PaginatedList.py", line 73, in _grow
    newElements = self._fetchNextPage()
  File "github/PaginatedList.py", line 198, in _fetchNextPage
    headers=self.__headers
  File "github/Requester.py", line 268, in requestJsonAndCheck
    return self.__check(*self.requestJson(verb, url, parameters, headers, input, self.__customConnection(url)))
  File "github/Requester.py", line 328, in requestJson
    return self.__requestEncode(cnx, verb, url, parameters, headers, input, encode)
  File "github/Requester.py", line 381, in __requestEncode
    status, responseHeaders, output = self.__requestRaw(cnx, verb, url, requestHeaders, encoded_input)
  File "github/Requester.py", line 405, in __requestRaw
    response = cnx.getresponse()
  File "tests/Framework.py", line 129, in getresponse
    self.__writeLine(output)
  File "tests/Framework.py", line 140, in __writeLine
    self.__file.write(line + '\n')
UnicodeEncodeError: 'ascii' codec can't encode character u'\xa0' in position 2437: ordinal not in range(128)

I haven't looked too closely at the code, but this looks related to the fact that this class treats the response body as ASCII and UTF-8 in different places.

There's another issue with the same line in Python 3. Files opened in binary mode expect to receive bytes and '\n' is a string.

[s-t-e-v-e-n-k]

Closed via #1254